面向目标综合识别的证据关联挖掘方法
|
摘要
针对海量异构数据中难以得到目标综合识别结果问题,提出了一种证据关联挖掘方法。该方法将证据与待证事实进行关联性量化分析,提炼出反映目标综合识别信息不同侧面的证据。首先,建立了证据历史信息分类模型和证据相关强度矩阵;然后,根据新证据信息对模型进行迭代更新,从而将识别信息源包含的数据和信息有效运用于证据理论,并得出目标综合识别的正确结论;最后,仿真试验表明该方法的可行性和有效性,可为多源异类信息融合处理提供参考。
Aimed at the problem about obtaining the integrated target identification result difficultly from the massive heterogeneous data,an evidence association mining method is proposed.The quantitative correlation analysis between the evidence and the fact to be verified is carried out,and the evidences reflecting different aspects of the integrated target identification information are extracted.Firstly,the classification model for the history evidence information and the related intensity matrix of the evidences are established.Then,the model is iteratively updated according to the new evidence information.The data and information contained in the identification information sources are applied to the evidence theory,and a correct conclusion of the integrated target identification is obtained.Finally,the simulation experiment shows its feasibility and validity.It can provide a reference to the multi-source heterogeneous information fusion.
Aimed at the problem about obtaining the integrated target identification result difficultly from the massive heterogeneous data,an evidence association mining method is proposed.The quantitative correlation analysis between the evidence and the fact to be verified is carried out,and the evidences reflecting different aspects of the integrated target identification information are extracted.Firstly,the classification model for the history evidence information and the related intensity matrix of the evidences are established.Then,the model is iteratively updated according to the new evidence information.The data and information contained in the identification information sources are applied to the evidence theory,and a correct conclusion of the integrated target identification is obtained.Finally,the simulation experiment shows its feasibility and validity.It can provide a reference to the multi-source heterogeneous information fusion.
引文
[1]王智玮.自动目标识别中全源数据挖掘技术应用[J].指挥信息系统与技术,2012,3(8):26-31.WANG Zhiwei.Full source data mining technology for automated target recognition[J].Command Information System and Technology,2012,3(8):26-31.(in Chinese)
[2]张燕君,龙呈.基于证据理论的目标识别方法[J].系统工程与电子技术,2013,35(12):2467-2470.ZHANG Yanjun,LONG Cheng.Target recognition method based on evidence theory[J].Systems Engineering and Electronics,2013,35(12):2467-2470.(in Chinese)
[3]许敬,王晓锋.基于贝叶斯概率的运动目标识别方法[J].南京理工大学学报,2013,37(1):76-81.XU Jing,WANG Xiaofeng.Recognition method of moving target using Bayesian probability theory[J].Journal of Nanjing University of Science and Technology,2013,37(1):76-81.(in Chinese)
[4]李丽荣,沈春林,王从庆.基于BP神经网络的空中目标识别方法[J].火力与指挥控制,2012,37(12):132-137.LI Lirong,SHEN Chunlin,WANG Congqing.Recognition method of aerial targets based on BP neural network[J].Fire Control&Command Control,2012,37(12):132-137.(in Chinese)
[5]苏伟,李为民,赵永.BP和D-S结合的多传感器协同目标识别推理机制[J].空军工程大学学报(自然科学版),2014,15(2):29-32.SU Wei,LI Weimin,ZHAO Yong.A Study of reasoning mechanism on multiple sensors cooperation with target identification based on BP neural network and DS evidence theory[J].Journal of Air Force Engineering University(Natural Science Edition),2014,15(2):29-32.(in Chinese)
[6]王惠娟,徐欣,李霄.雷达信号PRI抖动模式识别改进方法[J].指挥信息系统与技术,2016,7(2):69-72.WANG Huijuan,XU Xin,LI Xiao.Advanced identification method for radar signal PRI jitter pattern[J].Command Information System and Technology,2016,7(2):69-72.(in Chinese)
[7]王坚,史朝辉,郭新鹏,等.Mamdani模糊推理算法的直觉化扩展[J].计算机科学,2016,43(6A):44-46.WANG Jian,SHI Zhaohui,GUO Xinpeng,et al.Intuitionistic extension of Mamdani fuzzy Reasoning arithmetic[J].Computer Science,2016,43(6A):44-46.(in Chinese)
[8]刁联旺,王晓璇.目标识别证据相关性的统计度量[C]//第七届中国信息融合大会论文集.青岛:中国航空学会信息融合分会,2015:797-801.
[9]罗钧旻,李何,闫谦时.证据相关对推理结果可信度影响研究[J].西安工业大学学报,2013,33(3):183-186.LUO Junmin,LI He,YAN Qianshi.Research on influence of dependent evidence on reliability of reasoning results[J].Journal of Xi′an Technological University,2013,33(3):183-186.(in Chinese)
[2]张燕君,龙呈.基于证据理论的目标识别方法[J].系统工程与电子技术,2013,35(12):2467-2470.ZHANG Yanjun,LONG Cheng.Target recognition method based on evidence theory[J].Systems Engineering and Electronics,2013,35(12):2467-2470.(in Chinese)
[3]许敬,王晓锋.基于贝叶斯概率的运动目标识别方法[J].南京理工大学学报,2013,37(1):76-81.XU Jing,WANG Xiaofeng.Recognition method of moving target using Bayesian probability theory[J].Journal of Nanjing University of Science and Technology,2013,37(1):76-81.(in Chinese)
[4]李丽荣,沈春林,王从庆.基于BP神经网络的空中目标识别方法[J].火力与指挥控制,2012,37(12):132-137.LI Lirong,SHEN Chunlin,WANG Congqing.Recognition method of aerial targets based on BP neural network[J].Fire Control&Command Control,2012,37(12):132-137.(in Chinese)
[5]苏伟,李为民,赵永.BP和D-S结合的多传感器协同目标识别推理机制[J].空军工程大学学报(自然科学版),2014,15(2):29-32.SU Wei,LI Weimin,ZHAO Yong.A Study of reasoning mechanism on multiple sensors cooperation with target identification based on BP neural network and DS evidence theory[J].Journal of Air Force Engineering University(Natural Science Edition),2014,15(2):29-32.(in Chinese)
[6]王惠娟,徐欣,李霄.雷达信号PRI抖动模式识别改进方法[J].指挥信息系统与技术,2016,7(2):69-72.WANG Huijuan,XU Xin,LI Xiao.Advanced identification method for radar signal PRI jitter pattern[J].Command Information System and Technology,2016,7(2):69-72.(in Chinese)
[7]王坚,史朝辉,郭新鹏,等.Mamdani模糊推理算法的直觉化扩展[J].计算机科学,2016,43(6A):44-46.WANG Jian,SHI Zhaohui,GUO Xinpeng,et al.Intuitionistic extension of Mamdani fuzzy Reasoning arithmetic[J].Computer Science,2016,43(6A):44-46.(in Chinese)
[8]刁联旺,王晓璇.目标识别证据相关性的统计度量[C]//第七届中国信息融合大会论文集.青岛:中国航空学会信息融合分会,2015:797-801.
[9]罗钧旻,李何,闫谦时.证据相关对推理结果可信度影响研究[J].西安工业大学学报,2013,33(3):183-186.LUO Junmin,LI He,YAN Qianshi.Research on influence of dependent evidence on reliability of reasoning results[J].Journal of Xi′an Technological University,2013,33(3):183-186.(in Chinese)